Re: Traits.UI performance especially TableEditor

David C. Morrill schrieb:
> Franz-Josef Reinhardt wrote:
>> Hi,
>>
>> if I use the Traits.UI TableEditor with large Tables, eg. 3000 lines and 
>> 20 columns it takes almost 30 seconds from the configureTrais command 
>> till the TableEditor window appears. Same with the header sort.
>>
>> Any idea if there is a way to improve performance ?
>>
>>   
> 
> Try setting the TableEditor 'auto_size' trait to False (default is 
> True). The underlying wxWidgets code apparently has trouble doing the 
> 'auto_size' logic efficiently, which gets worse and worse as the number 
> of rows/cells increases. Let me know if that helps or not...
> 
> Dave Morrill

Hi Dave,

great! This did the trick. Thanks a lot. The time it takes from 
configure Traits to having the TableEditor displayed goes from 40 to 15 
seconds for a Table with more than 5000 rows and the haeder sort turns 
from unusable slow to very snappy.

I'm so impressed by traits. Everything my boss wants to be build in is 
already there, and I just have to turn it on. 

CU

Re: FW: mayavi question

Hi,

I'm ccing the mailing lists since this response is likely to be of
general interest.

>>>>> "Massimo" == DiPierro, Massimo <MDiPierro <at> cti.depaul.edu> writes:

    Massimo> First of all let me congratulate for developing mayavi. I
    Massimo> tried it and I am impressed.  I am the author of

Thanks for your compliments!

    Massimo> www.fermiqcd.net <http://www.fermiqcd.net/> and I am
    Massimo> starting a new project for visualization in lattice QCD.

Looks impressive!

    Massimo> I would very much like to use envisage and mayavi2. I am
    Massimo> new to vtk and mayavi so I am stuck with a problem and I
    Massimo> am hoping you could help me.

OK.

    Massimo> I am using the attached test program but my data source
    Massimo> (in the example heart.vtk) is periodically replaced by a
    Massimo> different file. How do I tell mayavi to periodically
    Massimo> reload the data from source without changing the other
    Massimo> parameters of the visualization? I am trying to implement
    Massimo> a real-time visualization of isosurfaces for the 3D Ising
    Massimo> model.

Where are the Traits User Preferences stored?

 From the ppt:

"The user preference values for a View are saved in a global “database” 
under the view’s ‘id’, so the View ‘id’ should also be unique across 
applications and views"

The preferences in my application are saved, but I can't find where.
I've seen a directory called "home" created before in the application 
folder, but I deleted it and now can not reproduce it. Preference saving 
still works though.

Cheers

Franz

_______________________________________________
Enthought-dev mailing list
Enthought-dev <at> mail.enthought.com
https://mail.enthought.com/mailman/listinfo/enthought-dev

Re: Traits.UI performance especially TableEditor

Franz-Josef Reinhardt wrote:
> Hi Dave,
>
> great! This did the trick. Thanks a lot. The time it takes from 
> configure Traits to having the TableEditor displayed goes from 40 to 15 
> seconds for a Table with more than 5000 rows and the haeder sort turns 
> from unusable slow to very snappy.
>   

Excellent!
> I'm so impressed by traits. Everything my boss wants to be build in is 
> already there, and I just have to turn it on. 
>
>   

That's nice to hear. Vielen Dank.

Dave Morrill

Re: [MayaVi-users] FW: mayavi question

> >>>>> "Massimo" == DiPierro, Massimo <MDiPierro@...> writes:

>     Massimo> www.fermiqcd.net <http://www.fermiqcd.net/> and I am
>     Massimo> starting a new project for visualization in lattice QCD.

I've attached some code which you may find useful as a starter.  Note
that it will NOT be good for you right away because it is quite old,
written for Mayavi1, plain VTK and Numeric.

But still, some of what's there may provide a first cut.  There are
classes in there to load Topological Densities in the format produced
by the version 5 MILC code in pure gauge configurations (from the
binary lattice files).  My use for it was to slice the field
configurations (I also did some fermion density plots using the same
code) in any direction and generate HTML tables of isosurfaces for
easy inspection of a large number of lattices.  I found it easier to
look at the data this way by dumping 'contact sheets' of the field
config than to manually load hundreds of configs.

If this of any use to you, do with it as you wish; I don't use this
code at all anymore.

Cheers,

f

_______________________________________________

Re: Install problem from SVN?

On 01/12/06, Prabhu Ramachandran <prabhu <at> aero.iitb.ac.in> wrote:
>     Angus> and then does nothing. There is a hint of an attempt to
>     Angus> create a window after the second line is printed, but it's
>     Angus> destroyed very quickly.
>     >>  Hmm, this is very strange.  Can you please try the following:

[snip]

>     >> 2. Can you run pyface/examples/scene.py and
>     >> pyface/examples/dscene.py
>
>     Angus> Yes it runs, but not correctly. Same outcome as for the
>     Angus> itvtk window, with garbage printed in the top-left
>     Angus> corner. Exiting via the menu worked okay though.
>
> IIUC, resizing the window has no effect, right?
>
> Can you try this:
>
>  cd tvtk/examples/plugin
>  python test.py
>
> See if this starts up fine.  Then try to open a new TVTK window by
> File->New->VTK Scene.
>
> If this does not work, the best way to find a fix is to use the
> pyface/examples/scene.py as a test case, then edit
> pyface/tvtk/scene.py the _create_control(...) method is the one which
> will need some hacking and experiment with a few print statements
> there to see where things are going wrong.

Re: Install problem from SVN?

On 12/1/06, Angus McMorland <amcmorl <at> gmail.com> wrote:

> Ultimately running the three lines:
>
> from enthought.tvtk.tools import ivtk
> ivtk.IVTKWithBrowser().open() # works okay
> ivtk.IVTKWithCrustAndBrowser().open() # throws up the weird window
>
> led me to think that the problem might be with the shell, since that's
> the difference between these two cases. I've had set ipython to run by
> default at startup, using:
>
> $PYTHONSTARTUP=$HOME/.pythonrc
>
> where .pythonrc is
>
> import IPython
> IPython.Shell.IPShell().mainloop(sys_exit=1)
>
> as listed in the ipython manual, (pdf page 67). However, this is the
> cause of the problem, and removing that startup file fixed the
> problem. It would appear that the threading behaviour of ipython and
> the shell(?) in mayavi2 was the problem. Now that we know, there isn't
> a problem for me per se, but it might be helpful to build in a check
> for this somehow.

Yes: you were starting the 'plain' ipython, and for interactive usage
of TVTK/M2, you need to handle the WX threading issues correctly.
IPython will do this, but you need to start it as

Re: [MayaVi-users] Install problem from SVN?

>>>>> "Angus" == Angus McMorland <amcmorl@...> writes:

    Angus> Eureka! I finally solved the problem, and this may be worth
    Angus> noting for others' benefit. From your hints, I started
    Angus> using ipython to run dig into the internals and find short
[...]
    Angus> Thanks for your help Prabhu. Without those hints on where
    Angus> to look I'd have been completely stuck.

Glad to know this is resolved.  Thanks for letting me know.  This is a
big relief.  I was quite concerned that despite the amount of effort
that went into the scene.py code, it did not work for you and that the
problem was not easy to debug.

I think the problem in this case is between the IPython interpreter
shell and the embedded shell in MayaVi2.  I am not sure it is a
threading issue as mentioned by Fernando since you were running
mayavi2 and its examples standalone applications.

I added those lines to my .python and am able to reproduce this
problem.  So it does look like an IPython/pyface-python-shell problem
to me and not a threading issue.

What does confuse me is that I thought PYTHONSTARTUP is only run for
interactive sessions and not for all sessions.  At least that is what
the man page suggests.

Anyway, glad to know that we have a solution to the problem.

cheers,

Wiki seems down

Upon access to https://svn.enthought.com/enthought I get the following
page:

"""
Traceback (most recent call last):
  File "/usr/lib/python2.4/site-packages/trac/web/main.py", line 356, in
dispatch_request
    dispatcher.dispatch(req)
  File "/usr/lib/python2.4/site-packages/trac/web/main.py", line 205, in
dispatch
    req.hdf = HDFWrapper(loadpaths=chrome.get_all_templates_dirs())
  File "/usr/lib/python2.4/site-packages/trac/web/chrome.py", line 263,
in get_all_templates_dirs
    dirs += provider.get_templates_dirs()
  File
"/var/www/trac/enthought/plugins/TracAccountManager-0.1.2.egg/acct_mgr/web_ui.py",
line 88, in get_templates_dirs
  File "/usr/lib/python2.4/site-packages/pkg_resources.py", line 799, in
resource_filename
    return get_provider(package_or_requirement).get_resource_filename(
  File "/usr/lib/python2.4/site-packages/pkg_resources.py", line 1229, in
get_resource_filename
    return self._extract_resource(manager, zip_path)
  File "/usr/lib/python2.4/site-packages/pkg_resources.py", line 1235, in
_extract_resource
    last = self._extract_resource(
  File "/usr/lib/python2.4/site-packages/pkg_resources.py", line 1249, in
_extract_resource
    real_path = manager.get_cache_path(
  File "/usr/lib/python2.4/site-packages/pkg_resources.py", line 880, in

Re: Wiki seems down

Gael,

This might be related to the reboot of the SVN machine that was  
performed this morning.  I'll forward the information on to our  
system admin.  Thanks for pointing it out!

-Peter